Power BI is turning 10! Tune in for a special live episode on July 24 with behind-the-scenes stories, product evolution highlights, and a sneak peek at what’s in store for the future.
Save the dateEnhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends August 31st. Request your voucher.
I am working on/am responsible for many different reports. Salesforce is our main data source, soon to be replaced by a Synapse database. Whenever I do a report, I spend literally days recreating the math behind the visuals I am using in any report, and I am thinking, there has to be a "best practices" to do this kind of thing. If I simply recreate the Power Query M code in Excel, it would be the same as using your training data to test your machine learning model. Should I simply create some SF reports for this? Should I write some Python code for this? Please share what has worked for you.
Hey @SDittmannFleet ,
What do you mean by "... recreating the math behind the visuals ..."?
I recommend creating a Power BI dataset that contains all the calculations, measures and calculated columns.
Then point your Power BI report to the Power BI dataset when creating your visuals instead of pointing to the data source.
This requires separating the data from the content and working with two (maybe more) pbix files, one pbix that is used to develop the data model and the second one (or many others) to develop to the conten, the data visualizations.
The content report is also called (thin report).
I always recommend separating the data with the content, even if it sometimes seems tedious, when you realize during content developement that a measure is missing, but in the long run the separation has a lot of advantages.
Look for thin report and Power BI shared datasets (https://learn.microsoft.com/en-us/power-bi/connect-data/service-datasets-across-workspaces?WT.mc_id=...).
Hopefuyll, this helps to tackle your challenge.
Regards,
Tom
That's not what I was referring to, but thank you for the tip, I'll check it out.
What I am referring to is, for example this: I create a visual based on a table with, say, some merges and other modifications done in Power Query. Of course I am assuming I did everything correctly, but if I want to double-check my logic to make sure the math lines up, it makes no sense to recreate the Power Query code in Excel and take it from there.
Check out the July 2025 Power BI update to learn about new features.
This is your chance to engage directly with the engineering team behind Fabric and Power BI. Share your experiences and shape the future.
User | Count |
---|---|
69 | |
68 | |
40 | |
29 | |
26 |
User | Count |
---|---|
86 | |
49 | |
45 | |
38 | |
37 |